Leibniz Institute for Prevention Research & Epidemiology – BIPS(莱布尼茨预防研究与流行病学研究所——BIPS);University of Bremen(不来梅大学);King’s College London(伦敦国王学院);University of Copenhagen(哥本哈根大学)
Despite the popularity of feature importance (FI) measures in interpretable machine learning, the statistical adequacy of these methods is rarely discussed. From a statistical perspective, a major distinction is between analyzing a variable's importance before and after adjusting for covariates - i.e., between $\textit{marginal}$ and $\textit{conditional}$ measures. Our work draws attention to this rarely acknowledged, yet crucial distinction and showcases its implications. Further, we reveal that for testing conditional FI, only few methods are available and practitioners have hitherto been severely restricted in method application due to mismatching data requirements. Most real-world data exhibits complex feature dependencies and incorporates both continuous and categorical data (mixed data). Both properties are oftentimes neglected by conditional FI measures. To fill this gap, we propose to combine the conditional predictive impact (CPI) framework with sequential knockoff sampling. The CPI enables conditional FI measurement that controls for any feature dependencies by sampling valid knockoffs - hence, generating synthetic data with similar statistical properties - for the data to be analyzed. Sequential knockoffs were deliberately designed to handle mixed data and thus allow us to extend the CPI approach to such datasets. We demonstrate through numerous simulations and a real-world example that our proposed workflow controls type I error, achieves high power and is in line with results given by other conditional FI measures, whereas marginal FI metrics result in misleading interpretations. Our findings highlight the necessity of developing statistically adequate, specialized methods for mixed data.
Conditional Feature Importance with Generative Modeling Using Adversarial Random Forests
Leibniz Institute for Prevention Research & Epidemiology – BIPS(莱布尼茨预防研究与流行病学研究所 – BIPS);Faculty of Mathematics and Computer Science, University of Bremen(不来梅大学数学与计算机科学学院);Berlin School of Economics and Law(柏林经济与法律学院);University of Copenhagen(哥本哈根大学)
Statistically Valid Variable Importance Assessment through Conditional Permutations
Inria(法国国家信息与自动化研究所);Universite Paris Saclay(巴黎萨克雷大学);CEA(法国原子能委员会);Roche Pharma Research and Early Development(罗氏制药研究与早期开发部);Neuroscience and Rare Diseases(神经科学与罕见病部门);Roche Innovation Center Basel(罗氏巴塞尔创新中心);F. Hoffmann–La Roche Ltd.(豪夫迈·罗氏有限公司)
Augmented Functional Random Forests: Classifier Construction and Unbiased Functional Principal Components Importance through Ad-Hoc Conditional Permutations
Faculty of Technological and Innovation Sciences(技术与创新科学学院);Faculty of Economics and Law(经济与法学院)
包括大量异质变量(例如计数、二进制、连续、偏斜连续等数据类型)的“混合数据”在不同领域中普遍存在,例如基因组学和蛋白质组学、成像遗传学、国家安全、社交网络和互联网广告。在统计建模这样的混合数据联合有限的努力,部分是因为缺乏计算上顺从的多变量分布,可以捕捉不同类型的这种混合变量之间的直接依赖关系。在本文中,我们解决这个问题,通过引入一类新的块有向马尔可夫随机场(BDMRFs)。使用Yang et al.(2012)的节点条件单变量指数族的基本构建块,我们引入了一类混合条件随机场分布,然后根据块定向无圈图以形成我们的块定向马尔可夫随机场(BDMRF)类。因此,BDMRF下的马尔可夫独立图结构具有有向边和无向边。我们引入这些分布存在的条件下,是规范化的,研究我们的模型的几个实例,并提出可扩展的惩罚条件似然估计与统计保证恢复底层网络结构。模拟以及从下一代测序表达数据和突变数据中学习混合基因组网络的应用证明了我们方法的多功能性。
Department of Statistics and Actuarial Science, The University of Hong Kong, Hong Kong SAR, China(香港大学统计与精算科学系);Musketeers Foundation Institute of Data Science, The University of Hong Kong, Hong Kong SAR, China(香港大学数据科学穆斯克特基金会研究所);Department of Statistics and Data Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA(卡内基梅隆大学统计与数据科学系);Computational Biology Department, Carnegie Mellon University, Pittsburgh, PA 15213, USA(卡内基梅隆大学计算生物学系);Machine Learning Department, Carnegie Mellon University, Pittsburgh, PA 15213, USA(卡内基梅隆大学机器学习系)